This proposal focuses on those drugs that are commonly utilized for the treatment of adult solid tumors, including both modern targeted oncology drugs, as well as cytotoxic agents. This includes studies of new candidate genes identified through recently completed genome-wide studies, both clinical and in the HapMap lymphoblastoid cell lines (LCLs). A central strategy of the proposal is the development of endophenotypes that are intermediate between clinical phenotypes, such as adverse events and measures of efficacy, and the genetic variants that affect their risk. All research projects are classified as primarily phenotype-centric, drug-centric, or gene-centric. The phenotype-centric studies will focus on markers of gene activity, such as the transcriptome, or potentially serum proteomic markers. The drug-centric studies will focus on drugs of importance for adult solid tumors. The gene-centric studies generally focus in a detailed way on candidate genes identified through these other studies. The overall goal is to identify functional relationships at the intersection of drug, gene, and phenotype, particularly those phenotypes that are affected by one or more functional polymorphisms when patients are treated with a specific anticancer drug (at a clinically relevant dose). Mathematically, this can be described by a multidimensional matrix (to be made publicly available through a PGScore database) of variation in a specific gene (or polymorphism) on the effects of a drug as measured by a specific phenotype. A consistent framework is utilized to address questions relevant to six interrelated Themes (Cytotoxicity, Transcriptome, Irinotecan, Angiogenesis, UGT, and EGFR), supported by two Platforms (Clinical Studies and Functional Studies) and four Cores (Management, Genetics-Informatics-Statistics, Liver, and LCL). Extensive collaborative clinical studies are proposed with CALGB, as well as other current PGRN Groups. Other ongoing collaborations will be continued, including studies of glucuronidation and LCLs. In addition, a new Network Resource is proposed. Cell Lines as a Resource for Pharmacogenomic Studies. PUBLIC HEALTH RELEVANCE: There is marked variability in pharmacological response to anticancer agents, which historically have been characterized by severe toxicity and inconsistent efficacy. This proposal aims to characterize the genomic basis for this variability through a series of interrelated studies addressing a breadth of modern and classical anticancer agents. This will include both laboratory and translational clinical studies.